Example of using GraphQL with InterSystems IRIS with Graphene, SQLAlchemy and Flask.

What are the goals of this example?

• Create a sample Employee, Department data model in InterSystems IRIS
• Implement a GraphQL server to query and modify data

What do you need to install?

• Git
• Docker (if you are using Windows, make sure you set your Docker installation to use “Linux containers”).
• Visual Studio Code + InterSystems ObjectScript Extension Pack
• Python

Setup

Run an InterSystems IRIS Community edition instance:

docker compose build
docker compose up -d

Create a Python virtual environment for this project:

# create a local venv environment
python3 -m venv .venv
activate venv
source .venv/bin/activate
install dependencies
pip3 install -r requirements.txt

Create some sample data

Open an interactive Python session and run a method to initialize your data model and insert some data.

cd graphql-server
python

from models import db_session, engine, Base, Department, Employee
# create tables
Base.metadata.create_all(bind=engine)
# create some data
from app import init_data
init_data()

Run GraphQL server

There is a GraphQL server implementation ready for you to test:

• graphql-server/models.py - defines the database schema. It includes the models for Employee and Department.
• graphql-server/schema.py - GraphQL schema definition using Graphene. It includes queries and mutations for creating data.
• graphql-server/schema-relay.py - Another GraphQL schema definition using Graphene but using a relay approach that supports pagination.
• graphql-server/app.py - sets up the Flask application and /graphql and /graphql-relay endpoints for both schemas (relay and not relay).

cd graphql-server
python app.py

Test your GraphQL implementation

You can import collection_iris-graphql-demo.json in Postman or ThunderClient for VS Code and check the output.

There is also a GraphiQL UI available at http://localhost:5000/graphql